Collaborative filtering algorithm based on mutual information

As information spaces such as the WWW grow ever larger, the need for tools to help users find high quality reliable information quickly and easily becomes ever more acute. Collaborative filtering (CF) based recommender systems have emerged in response to these problems. Collaborative filtering is a popular technique for reducing information overload and has seen considerable successes in many area. In order to further improve the accuracy of the Collaborative filtering, a new approach for the collaborative filtering algorithms is proposed using mutual information. The new method is based on a simultaneous approach to feature weighting and relevant instance selection. The proposed methods are evaluated on the well-known EachMovie dataset and the experimental results demonstrate a significant improvement in accuracy and efficiency.